Is Agentive Certified recognised by EU regulators?
No, and we are deliberate about saying so. Agentive Certified is a private quality mark structured against Regulation (EU) 2024/1689 and aligned with ISO/IEC 42001:2023. It is not a regulatory pre-clearance, an EU recognition, or a substitute for accredited third-party conformity assessment under Article 43 (mandatory for some high-risk systems). What it is: preparation, evidence, and discipline. If a competent authority audits you tomorrow, this is the binder you hand them. That is the honest, useful claim — and it's the same posture SOC 2 took to break into the information-security market.
How does this differ from ISO/IEC 42001?
ISO/IEC 42001:2023 is a global, jurisdiction-agnostic AI management system standard. It tells you how to structure an AI governance programme; it does not tell you what the EU AI Act requires of you specifically. The European Commission has expressly stated that ISO/IEC 42001:2023 is not aligned with the AI Act's required quality management system, and the AI Act does not confer presumption of conformity on ISO/IEC 42001 conformance. Agentive Certified is the EU-AI-Act-specific overlay on top of that logic: twelve control domains, each citing the relevant Article(s), with the literacy programme (Article 4) built in. Many clients run both: ISO 42001 as the global frame, Agentive Certified as the EU-jurisdiction operational layer.
Which Irish regulator supervises Article 4 for my company?
It depends on your sector. Ireland uses a distributed enforcement model under S.I. No. 366/2025 with 15 designated National Competent Authorities, coordinated by the future AI Office of Ireland. A financial-services firm answers to the Central Bank of Ireland. A healthcare provider answers to the HPRA. A media or online-platform business answers to Coimisiún na Meán. A telecoms operator answers to ComReg. A general professional-services SME using AI tools sits in the residual market-surveillance route. Where personal data is processed, the Data Protection Commission is concurrently relevant. The Agentive evidence pack is structured to satisfy the question any of these regulators is most likely to ask first: who was trained, on what, against which systems in your register.
Are there Irish-specific Article 4 fines yet?
Not yet. Article 4 has no harmonised fine tier in Article 99 itself; Member States must establish their own residual penalty regime under Article 99(1). In Ireland, those rules are being legislated through the Regulation of Artificial Intelligence Bill 2026, of which DETE published a General Scheme on 4 February 2026. Until the Bill is enacted, no Irish AI-Act-specific administrative fine for Article 4 exists. The commercial risk is downstream: untrained staff deploying systems that fall within Annex III high-risk or Article 50 transparency duties, both of which sit in the 3% / €15M tier under Article 99(4) once those obligations apply. The Agentive programme is designed to keep you out of that downstream exposure, irrespective of where the standalone Article 4 fine lands in Irish law.
Does Article 4 apply to my company?
If your company provides or deploys an AI system within the scope of the Regulation, yes. Article 4 has no SME carve-out and no minimum-headcount threshold. The Regulation itself has scope exclusions in Article 2 (e.g. military and national-security uses, pure scientific R&D, personal non-professional use). For ordinary business use of AI tools by staff (ChatGPT, Copilot, AI-powered SaaS), the duty applies.
What does "sufficient literacy" actually mean?
The Act says, verbatim: "Providers and deployers of AI systems shall take measures to ensure, to their best extent, a sufficient level of AI literacy of their staff and other persons dealing with the operation and use of AI systems on their behalf, taking into account their technical knowledge, experience, education and training and the context the AI systems are to be used in, and considering the persons or groups of persons on whom the AI systems are to be used." (Article 4, Regulation (EU) 2024/1689). In plain English: literacy is role-proportional, context-dependent, and audience-aware. Our programme is designed to evidence each of those three legs, structured around the European AI Office's living guidance plus the consensus reading of Articles 3, 4, 13, 14 and 50.
How do you handle ongoing changes?
Two ways. First, the curriculum is versioned: we update quarterly when the AI Office publishes new guidance. Second, our optional Compliance Subscription keeps your register fresh, your evidence pack current, and pushes a regulator-update bulletin to your inbox each month.
Why founder-led delivery?
Because the alternative is a junior trainer reading a slide deck. Victor del Rosal and Joe Wilde personally lead every engagement — kick-off, the workshops that matter, and the evidence sign-off. Capacity is finite, which is part of the price logic.
What if a regulator audits us next year?
Hand them the evidence pack. It's structured to answer the questions a competent authority is likely to ask: who was trained, when, on what curriculum, with what assessment outcome, against which systems in your register. We design it to be the document you'd want to receive if you were the auditor.
How does pricing work?
Open the calculator. It returns an indicative midpoint with ±15% range, valid 14 days. Bigger orgs (over ~100 staff to certify) route to a 20-min scoping call. Either way, the final fee is fixed in writing before kick-off.
